Cheibub-Latin America Presidentialism PDF

CHICAGO
PUBLIC LAW AND LEGAL THEORY WORKING PAPER NO. 361

LATIN AMERICAN PRESIDENTIALISM IN COMPARATIVE AND
HISTORICAL PERSPECTIVE

José Antonio Cheibub, Zachary Elkins, and Tom Ginsburg

THE LAW SCHOOL
THE UNIVERSITY OF CHICAGO

September 2011

This paper can be downloaded without charge at the Public Law and Legal Theory Working Paper
Series: http://www.law.uchicago.edu/academics/publiclaw/index.html
and The Social Science Research Network Electronic Paper Collection.
Electronic copy available at: http://ssrn.com/abstract=1899637

Latin American Presidentialism in Comparative and
Historical Perspective
José Antonio Cheibub,* Zachary Elkins,**

& Tom Ginsburg***
Since the time of Aristotle, comparative scholars have developed
various alternative typologies to classify constitutional systems.1 One
paradigmatic scheme focuses on executive–legislative relations. Some
systems,2 we are told, are ―presidential,‖ in which a directly elected president
serves a fixed term as both head of state and head of government. Others are
―parliamentary,‖ in which a legislative majority determines who will lead the
government and for how long. A third model combines features of the two
and is called ―semi-presidential.‖3
Each of these models of ―government type‖ has an archetype: The
United States is seen as the quintessential presidential system,4 the United
Kingdom as the parliamentary model,5 and France as the semi-presidential
model.6 The models are also seen as systemic, in that each implies a certain
institutional configuration. So, presidential systems are thought to include a
host of features (e.g., an executive veto) that are not typically found in
* Professor of Political Science, University of Illinois at Urbana-Champaign.

** Associate Professor of Government, University of Texas at Austin.
***Professor, University of Chicago Law School. The authors thank John Ferejohn and
Jonathan Hartlyn for excellent comments, as well as the editors of the Texas Law Review.
1. See, e.g., Roberto Gargarella, Towards a Typology of Latin American Constitutionalism,
1810–60, 39 LATIN AM. RES. REV. 141, 142 (2004) (characterizing early Latin American
constitutions as conservative, majoritarian, and liberal).
2. Patterns of executive–legislative relations are sometimes referred to as ―forms‖ or ―systems‖
of government. Scholars sometimes even summarize constitutions on the basis of these differences
(e.g., labeling constitutions as either presidential or parliamentary), an indicator of the centrality of
these features to constitutional structure more generally.
3. There is no consensus among scholars on the definition of forms of government, particularly
with respect to semi-presidentialism. For a thorough review of the concept of semi-presidentialism
and the definitional controversies therein, see Robert Elgie, The Politics of Semi-Presidentialism, in
SEMI-PRESIDENTIALISM IN EUROPE 1, 1–14 (Robert Elgie ed., 1999).
4. See Keith E. Whittington, Yet Another Constitutional Crisis?, 43 WM. & MARY L. REV.
2093, 2127 (2002) (―Presidential systems are defined by the separate elections of the legislature and
the head of the government (the president) and by the fixed term of the president. The United States
is the classic example of such a system, and indeed is the longest enduring democratic presidential
system in the world.‖ (footnotes omitted)).
5. See Walter F. Murphy, Designing a Constitution: Of Architects and Builders, 87 TEXAS L.
REV. 1303, 1317 (2009) (―The classic model for representative democracy is the British
parliamentary system from 1867 through the United Kingdom‘s joining the European Union.‖).
6. Elgie, supra note 3, at 2–3 (―[In 1970], according to Duverger, the list of semi-presidential
regimes comprised three Western democracies, Austria, Finland, and France . . . .‖).

2 Texas Law Review [Vol. 89:XXXX
parliamentary systems.7 In turn, heads of government in parliamentary

systems are thought to be vested with powers that their counterparts in presi-
dential systems do not have (e.g., decree power or legislative initiative).8
In a recent paper, we have argued that the conventional categories are
not systemic in this sense.9 Indeed, we found that stereotypes regarding
presidentialism and parliamentarism are just that—stereotypes. When we
looked at the distribution of several presumably systemic features, we found
that only one of them could be described as a distinctive feature of one sys-
tem or the other.10 Our findings are captured by the following empirical
insight from our analysis: The century or region in which a constitution was
written is a better predictor of institutional similarity (with respect to the
studied features) than is its classification as presidential, parliamentary, or
semi-presidential.11 The categories have a degree of internal cohesion, but
not nearly as much as one would expect for categories that are thought to
represent a fundamental and guiding set of choices for constitutional
designers, especially given the tremendous scholarly literature built around
them.12
7. Thomas Weishing Huang, The President Refuses to Cohabit: Semi-Presidentialism in

Taiwan, 15 PAC. RIM L. & POL‘Y J. 375, 380 n.29 (2006) (―[P]residentialists argue that the
existence of presidential independent powers, particularly the power to veto legislation, makes it a
presidential system.‖).
8. See Charles Wallace Collins, Constitutional Aspects of a National Budget System, 25 YALE
L.J. 376, 376 (1916) (―[In] the parliamentary system of government[,] the executive possesses the
right of legislative initiative, actively participates in legislation on the floor of the legislature, and
through the prime minister as party leader controls the legislative output.‖).
9. José Antonio Cheibub, Zachary Elkins & Tom Ginsburg, Beyond Presidentialism and
Parliamentarism: On the Hybridization of Constitutional Form 20 (Feb. 28, 2010) (unpublished
manuscript) (on file with the Texas Law Review). Other scholars have made similar arguments.
See generally Richard Albert, The Fusion of Presidentialism and Parliamentarism, 57 AM. J. COMP.
L. 531 (2009) (emphasizing functional similarities between the two types); André Krouwel,
Measuring Presidentialism and Parliamentarism: An Application to East European Countries, 38
ACTA POLITICA 333 (2003) (arguing for the analysis of Eastern and Central European nations on a
continuum of presidentialism rather than on a categorical basis).
10. Cheibub, Elkins & Ginsburg, supra note 9, at 26.
11. Id. at 25.
12. See, e.g., TORSTEN PERSSON & GUIDO TABELLINI, THE ECONOMIC EFFECTS OF
CONSTITUTIONS (2003) (discussing the empirical correlations of economic effects with different
forms of government); Juan J. Linz, Presidential or Parliamentary Democracy: Does It Make a
Difference?, in 1 THE FAILURE OF PRESIDENTIAL DEMOCRACY 3 (Juan J. Linz & Arturo
Valenzuela eds., 1994) (discussing the role that parliamentary and presidential political institutions
play in shaping democratic decisions); Scott Mainwaring, Presidentialism, Multipartism and
Democracy: The Difficult Combination, 26 COMP. POL. STUD. 198, 222 (1993) (―[T]he combination
of presidential government and a multiparty system is problematic.‖); Matthew Soberg Shugart &
Scott Mainwaring, Presidentialism and Democracy in Latin America: Rethinking the Terms of the
Debate, in PRESIDENTIALISM AND DEMOCRACY IN LATIN AMERICA 12 (Scott Mainwaring &
Matthew Soberg Shugart eds., 1997) [hereinafter Shugart & Mainwaring, Rethinking the Terms of
the Debate] (defining presidential democracy in contrast to parliamentarism and analyzing the
performance and effectiveness of presidential regimes); Alfred Stepan & Cindy Skach,
Constitutional Frameworks and Democratic Consolidation: Parliamentarism Versus
Presidentialism, 46 WORLD POL. 1 (1993) (arguing that parliamentarism is a more supportive
constitutional framework for consolidating democracy than presidentialism); see also Albert, supra

2011] Latin American Presidentialism 3
One implication of this insight is that scholars need to explore

alternative conceptualizations of executive–legislative relations. The
distinction between assembly-confidence governments13 and directly elected,
fixed-term governments represents an important dimension—but only one
dimension—in a clearly multidimensional conceptual space. In this Article,
we explore this multidimensionality in the context of Latin America. Latin
America provides a useful context for exploring variety within constitutional
forms because of its monotypic history: since the emergence of the first
independent states in the region early in the nineteenth century, the region
has been dominated by the presidential model.14 Indeed, of the former
Spanish and Portuguese colonies in the Americas, the only country that
adopted a lasting nonpresidential constitution was Brazil, from 1824 to
1891.15 This apparent uniformity presents an opportunity to examine internal
diversity within a single overarching category of presidential systems.
When we explore the architecture of executive–legislative relations in
Latin America, it becomes clear that region matters as much as government
type in predicting the distribution of constitutional provisions. Latin
American presidentialism, while sharing a fair number of features with the
U.S. archetype, is very much its own breed. What appears to distinguish the
Latin American variety is a high degree of what we might summarize as
executive lawmaking powers. Specifically, Latin American constitutions are
uniquely inclined to empower presidents to decree laws, initiate legislative
proposals, and exert powers in emergency conditions. None of these powers
is stereotypical of presidentialism—indeed, some of them are thought to be
elective attributes of parliamentarism. Yet they are undeniably important
powers with potentially significant consequences for political stability and
the quality of democracy. Indeed, it may well be that the dimension of
executive lawmaking authority is found to be as important as the executive-
selection features that distinguish presidential and parliamentary
constitutions.
I. The Shadow of the U.S. Constitution

We begin with a historical elaboration of the influence of the U.S.
Constitution on Latin American constitutionalism as a way of orienting the
discussion. The influence of the U.S. Constitution in Latin America was
undoubtedly significant in the early nineteenth century. Among others,
note 9, at 531 (―Parliamentarism and presidentialism are commonly, and correctly, set in opposition
as distinguishable systems of governance that exhibit distinguishable structural features.‖).
13. See JOSÉ ANTONIO CHEIBUB, PRESIDENTIALISM, PARLIAMENTARISM, AND DEMOCRACY
36–37 (2007) (defining assembly confidence as a political system in which the government‘s
authority is constrained by the continued approval and confidence of the legislative assembly).
14. See infra notes 32–42 and accompanying text.
15. See Keith S. Rosenn, Separation of Powers in Brazil, 47 DUQ. L. REV. 839, 840–42 (2009)
(describing the legislative power under the 1824 constitution as parliamentary and noting that the
1824 constitution was Brazil‘s ―most enduring‖).

Venezuela‘s constitution of 1811, Mexico‘s of 1824, Argentina‘s of 1826,

and Ecuador‘s of 1830 drew significantly on the American model.16 Even
when not adopted, American institutions were part of the mix of models
considered. Argentina‘s constitution of 1853 was particularly close to the
U.S. model, so much so that Argentinian judges routinely drew on U.S. con-
stitutional jurisprudence in interpreting their own constitution for more than a
century.17 Indeed, there was so much borrowing that the great liberator
Simón Bolívar was ―moved to condemn the ‗craze for imitation.‘‖18
To be sure, the U.S. model was only one of several on offer. Latin
American elites were fully acquainted with enlightenment thought and drew
on eclectic sources, including French and British thought and, notably, the
1812 Constitution of Cádiz, the embodiment of Spanish liberalism.19
Nevertheless, several features of the U.S. model were particularly attractive.
Federalism was the leading example, as it helped accommodate traditions of
regional and municipal autonomy within the Spanish empire and served as an
attractive model for rural elites fearful of domination by urban centers.20
Venezuela‘s 1811 document drew directly and self-consciously on the United
States‘ federal model.21 Federalist thought was even influential in countries
where it was not sustained, such as Chile.22 As various independent states
sought to combine into larger entities, federalism was a natural model. The
Central American Federation, which encompassed much of that region from
1823 to 1840, was explicitly federal and drawn from the U.S. model.23 Gran
Colombia, which encompassed the territory of today‘s Colombia, Venezuela,
Panama, and Ecuador from 1819 to 1831, was also a federal republic.24
Today, Argentina, Brazil, Mexico, and Venezuela remain federal states.25
16. Donald L. Horowitz, The Federalist Abroad in the World, in THE FEDERALIST PAPERS 502,
505 (Ian Shapiro ed., 2009); Robert J. Kolesar, North American Constitutionalism and Spanish
America: “A Special Lock Ordered by Catalogue, Which Arrived with the Wrong Instructions and
No Keys”?, in AMERICAN CONSTITUTIONALISM ABROAD 41, 53–54 (George Athan Billias ed.,
1990); Miguel Schor, Constitutionalism Through the Looking Glass of Latin America, 41 TEX.
INT‘L L.J. 1, 15 (2006).
17. Kolesar, supra note 16, at 53–56. For a thorough discussion of this topic, see
JONATHAN M. MILLER, BORROWING A CONSTITUTION: THE U.S. CONSTITUTION IN ARGENTINA
AND THE HEYDAY OF THE ARGENTINE SUPREME COURT (1853–1930) (forthcoming 2012).
18. Horowitz, supra note 16, at 505 (quoting BERNARD BAILYN, TO BEGIN THE WORLD ANEW
146 (2003)).
19. Kolesar, supra note 16, at 42–43; see also Zachary Elkins, Diffusion and the
Constitutionalization of Europe, 43 COMP. POL. STUD. 969, 984 (2010) (comparing the influence of
different constitutional models on constitutions in Europe to that process in Latin America).
20. Kolesar, supra note 16, at 43–44.
21. Id. at 43.
22. See id. at 51 (―[D]uring the early years of independence, . . . North American constitutional
principles came to be closely associated with federalism in Chile.‖).
23. Horowitz, supra note 16, at 505. For background on the Central American Federation, see
LYNN V. FOSTER, A BRIEF HISTORY OF CENTRAL AMERICA 134–51 (2000).
24. See DAVID BUSHNELL, THE MAKING OF MODERN COLOMBIA: A NATION IN SPITE OF
ITSELF 51–52 (1993) (describing the process by which Gran Colombia became a federal republic).
25. ZACHARY ELKINS ET AL., THE ENDURANCE OF NATIONAL CONSTITUTIONS 26 (2009).
Judicial review was also an American export. Many constitutions in the

region adopted explicit provisions empowering the courts to exercise the
power of judicial review, a power only implicit in the United States
Constitution.26 To be sure, there were limitations on its exercise. In the
widely borrowed Mexican institution of amparo, courts could not strike a
statute even if they found it unconstitutional; they could only correct its
application in particular cases.27 It is not surprising that, until the emergence
of democracy in the late twentieth century, Latin American courts were
hardly vigorous in using their powers of review;28 but the similarity in
constitutional form nevertheless set the region apart from other parts of the
world.
Other institutions were adopted but then discarded. The right to bear
arms existed in many early Latin American constitutions, but by the turn of
the twentieth century it had almost been eliminated.29 The electoral college
was influential in early constitutions and survived perhaps longer than it
should have, remaining in place in Argentina until 1995.30 These modifica-
tions over time may have reflected a process of updating or modernization, as
nations experimented with institutions and found that some worked while
others did not.31
One of the major borrowings was, of course, the presidency. After a
nonnegligible period of experimentation, Latin American countries stabilized
under presidential constitutions in the nineteenth century.32 The choice of a
presidential form of government may perhaps be accounted for simply by the
fact that it was a model that was available. At independence, Latin American
countries were struggling with the same fundamental problem with which
leaders of the newly independent United States struggled after 1776: how to
constitute executive authority in a context where the monarch was no longer
the ruler. Parliamentary government had not yet been codified as such and
was in the process of emerging out of recently constitutionalized European
monarchies.33 Parliamentary constitutions in Europe emerged after a gradual
26. See Keith S. Rosenn, Judicial Review in Latin America, 31 OHIO ST. L.J. 785, 785 (1974)
(―A region of chronic political instability and short-lived constitutions with a civil law tradition
would appear most infertile soil for the seeds of Marbury v. Madison to take root. Yet all of the
Latin American republics, with the exception of the Dominican Republic, provide for some form of
judicial review.‖ (footnotes omitted) (citing Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803))).
27. Id. at 796.
28. See id. at 791–808 (surveying the historical development of mechanisms for raising
constitutional questions within Latin American countries).
29. ELKINS ET AL., supra note 25, at 27 & fig.2.2.
30. Horowitz, supra note 16, at 505.
31. ELKINS ET AL., supra note 25, at 28.
32. CHEIBUB, supra note 13, at 150.
33. See Adam Przeworski et al., The Origins of Parliamentary Responsibility, in COMPARATIVE
CONSTITUTIONAL DESIGN (Tom Ginsburg ed., forthcoming 2011) (manuscript at 150, 156–61) (on
file with authors) (detailing the rise of parliamentary monarchies in Europe and tracing the ―shift of
the power to appoint governments from the crown to elected assemblies‖).
period of negotiation between the monarch and the nobles, in which the par-
liament ultimately displaced the monarchy as the center of effective
governance.34 In Latin America, by contrast, initial governments, whether
revolutionary or not, emerged from a system of monarchy in which a single
individual sat at the center of the political system.35 Even Simón Bolívar,
who cloaked his critique of federalism in a general reaction to borrowing
from North America, was an admirer of the presidency as a model of a
nonhereditary yet strong executive.36 Thus, the drafters of presidential
constitutions in nineteenth-century Latin America did not choose between the
presidential and parliamentary models available today, but between a monar-
chy (headed by a hereditary leader) and a republic (headed by leaders with no
claim to heredity).37
At that time, the United States represented the most successful republic
and one that had emerged under similar circumstances.38 France, after all,
had not yet been able to settle upon a stable and coherent model of republi-
can government.39 Adoption of a presidential formula was perhaps a
foregone conclusion.
At the same time, the initial choice of presidentialism does not
necessarily explain the endurance of that model. After all, many other
institutions were discarded over time through processes of amendment and
constitutional replacement. There are reasons, however, to suppose that
basic constitutional frameworks—such as the one embodied in the proce-
dures for the selection of the executive—are subject to strong inertial factors.
These broad institutions structure the expectations of the actors operating
under them and, in order for them to be changed, actors must be willing to
leap into the unknown. At the same time, constitutions serve as focal points
and are rarely written on a blank slate; previous documents often serve as a
template, even if changes are made to address issues identified as leading to
crisis in prior systems of government. Thus, in spite of frequent
34. See id. at 156–57 (noting that although ―[c]onstitutional monarchs were chief executives . . .
who governed with the advice and consent of their ministers[,] . . . there were many instances in
which parliamentary majorities forced monarchs to dismiss or accept governments against their
will‖ and that ―[t]he power of the parliaments stemmed from their control over legislation,
particularly budgets‖).
35. See id. at 175–76 (describing the development of constitutions in Portugal and Spain and
noting the central role of the monarch in each nation).
36. Kolesar, supra note 16, at 50.
37. CHEIBUB, supra note 13, at 151.
38. See Kolesar, supra note 16, at 44 (noting that the ―social and economic success of the
United States‖ prompted Latin American drafters to consider the principles embodied in the
Constitution); id. at 58 (―North American constitutionalism was influential precisely because it
embodied values and addressed needs shared by many [Latin] Americans.‖).
39. The first stable republican government in France emerged in 1875. See ELKINS ET AL.,
supra note 25, at 169 (―The constitution that emerged [in 1875] was a compromise that combined a
strong chamber of deputies elected by universal suffrage and an upper house composed of senators
selected by local notables or appointed for life terms. Combining both popular and conservative
impulses, these institutions nevertheless facilitated the dominance of republicanism . . . .‖).
constitutional replacements in Latin America40 that in theory would have

provided many opportunities to reconsider presidentialism, and in spite of the
existence of explicit and vigorous attempts to reform it,41 the presidential
form of government has survived and shows no signs that it will be aban-
doned any time soon.42
This does not mean, however, that when adopted by Latin American
countries, presidentialism was taken as a package deal. If it is true that Latin
American countries borrowed the presidential solution from the United
States, it is not correct to assume that they also borrowed the set of ancillary
institutions that structure the powers of the president and the specific ways in
which the president is to interact with the legislature. Even if some such
ancillary institutions were borrowed, they too might evolve over time to
create new variants of presidentialism that bear little resemblance to the U.S.
model. Finally, it could be the case that presidential systems are sufficiently
internally diverse such that the overall category is hiding important variation.
These are empirical questions that have not, to our knowledge, been system-
atically examined before. It is our purpose in this Article to do just that in
the context of Latin America. We approach the issue of government type by
examining several internal features that are seen to be essential components
of presidential systems. It is to this issue that we now turn.
II. The U.S. Constitution as the Archetype of Presidentialism

As we stipulated above, scholars who focus on the study of political
systems see presidential and parliamentary types as representing systems of
institutions. As put by Moe and Caldwell, ―Presidential and parliamentary
systems come with their own baggage. They are package deals.‖43 The pre-
cise list of attributes that is supposed to be associated with each system is
subject to some variation. Some of these attributes may be accidental, while
others may follow from the logic of presidential governance. Tsebelis, for
example, asserts that ―[i]n parliamentary systems the executive (government)
controls the agenda, and the legislature (parliament) accepts or rejects
proposals, while in presidential systems the legislature makes the proposals
40. See infra Appendix A.

41. See Scott Mainwaring & Matthew Soberg Shugart, Introduction to PRESIDENTIALISM AND
DEMOCRACY IN LATIN AMERICA, supra note 12, at 1, 2 (discussing the efforts of Brazil, Argentina,
Colombia, Chile, and Bolivia to shift away from a presidential form of government).
42. We note, of course, that some prominent and recent episodes of constitutional design took
up the issue of presidentialism versus parliamentarism, including Argentina in 1993 and Brazil in
1988. For a discussion of frequent replacement of constitutions by these two countries and others,
see ELKINS ET AL., supra note 25, at 26.
43. Terry M. Moe & Michael Caldwell, The Institutional Foundations of Democratic
Government: A Comparison of Presidential and Parliamentary Systems, 150 J. INSTITUTIONAL &
THEORETICAL ECON. 171, 172 (1994) (Ger.).
and the executive (president) signs or vetoes them.‖44 Others emphasize the
following as key attributes of political systems: decree power,45 emergency
rule,46 veto power,47 legislative initiative,48 cabinet formation,49 and the
power to dissolve the assembly.50 This last feature is so closely linked with
parliamentarism that some even include it as a defining attribute.51
The United States Constitution represents the archetypical presidential
system in the sense that it is the model that represents, often implicitly,
discussions of separation-of-powers systems. What defines the U.S.
Constitution as presidential is that the executive is popularly elected and does
not need the confidence of the legislature in order to remain in office.52
Other features of the U.S. presidential system may or may not be unique and
include the following: First, the U.S. President is unable to dissolve the
assembly.53 Second, the President lacks explicit lawmaking powers and has
44. George Tsebelis, Decision Making in Political Systems: Veto Players in Presidentialism,
Parliamentarism, Multicameralism and Multipartyism, 25 BRIT. J. POL. SCI. 289, 325 (1995).
45. See, e.g., Lee Kendall Metcalf, Measuring Presidential Power, 33 COMP. POL. STUD. 660,
663 tbl.1 (2000) (citing Timothy Frye, A Politics of Institutional Choice: Post-Communist
Presidencies, 30 COMP. POL. STUD. 523 (1997)) (including the power to ―[i]ssue[] decrees in non
emergencies‖ among Frye‘s twenty-seven listed presidential powers).
46. See, e.g., BRIAN LOVEMAN, THE CONSTITUTION OF TYRANNY: REGIMES OF EXCEPTION IN
SPANISH AMERICA 5–6 (1993) (―Latin American constitutions almost always included provisions
for ‗emergency powers,‘ . . . to be used in times of internal strife or external threat.‖).
47. See, e.g., Albert, supra note 9, at 542–43 (characterizing the presidential veto in the U.S.
Constitution as a legislative power).
48. See, e.g., José Antonio Cheibub, Making Presidential and Semi-presidential Constitutions
Work, 87 TEXAS L. REV. 1375, 1386–88 (2009) (noting that ―[a]lmost all presidential constitutions
give some legislative powers to the presidency,‖ including the ―exclusive power to introduce
legislation in some specified areas‖).
49. See, e.g., Metcalf, supra note 45, at 660, 663 tbl.1 (citing Frye, supra note 45) (including
the power to ―[a]ppoint[] senior officers‖ among Frye‘s twenty-seven listed presidential powers).
50. See, e.g., Krouwel, supra note 9, at 339, 342–45 (distinguishing presidential, semi-
presidential, and parliamentary systems on several dimensions, including the ability of various
political actors to dissolve the legislature).
51. See, e.g., Stepan & Skach, supra note 12, at 3 (including the executive‘s ability to dissolve
the legislature as one of two ―fundamental characteristics‖ of a ―pure parliamentary regime‖).
52. See U.S. CONST. art. II, § 1, cl. 1–3, amended by U.S. CONST. amend. XII (providing for a
fixed presidential term of four years and popular election of the president through the electoral
college).
53. See U.S. CONST. art. II, amended by U.S. CONST. amend. XII & XXV (defining the powers
of the executive, which do not include the power to dissolve Congress). Although the power to
dissolve the assembly is often considered to be an essential, even defining, feature of the separation
of powers system, we do not take this position. Dissolution powers originated in monarchies and
are compatible today with all forms of democratic constitutions. Just as there are presidential
constitutions that allow dissolution under certain circumstances, there are parliamentary ones that
do not. Compare CONSTITUCIÓN DE LA REPÚBLICA DEL ECUADOR [ECUADOR CONST.] 2008,
art. 148 (listing certain circumstances under which the president can dissolve the national
assembly), with Przeworski et al., supra note 33, at 158 (noting Norway as an exception to the
general rule that, in countries operating under a constitutional monarchy, kings can dissolve
parliaments).
no constitutional power of executive decree.54 Third, the President has

formal, albeit modest, emergency powers.55 Fourth, the President lacks the
formal ability to initiate legislation but has the power to veto legislation, even
if the veto can be overridden.56 Fifth, the President has the ability to dismiss
the cabinet without direct legislative involvement.57 Sixth, the President has
the power of pardon.58 Seventh, the President is subject to explicit term
limitations, although those limits were not formalized until 1951.59 Eighth,
the legislature has explicit powers of oversight over the President.60 These
presumably elective features of presidentialism are the focus of our inquiry.
We are aware that some of these features are not necessarily descriptive
of how the U.S. presidential system works de facto. Some of the constraints
presidents face might result from informal rather than formal limitations. For
example, the two-term limit for presidents had long been observed before it
was formalized by the Twenty-Second Amendment.61 If presidents are for-
mally prevented from setting the legislative agenda, it is not hard for them to
find willing legislators to sponsor their bills. On the other hand, if the formal
constitution provides for a president devoid of strong constitutional powers,
in practice the U.S. President hardly seems weak (or, at least, seems to have
gained strength over the years). The expansion in the scope and frequency of
executive orders62 and the ongoing debate about executive powers in times of
war attest to this perception.63 Our goal, however, is to investigate whether
the constitutional documents crafted in Latin America correspond to the
54. See U.S. CONST. art. I, § 1 (―All legislative Powers herein granted shall be vested in a
Congress of the United States, which shall consist of a Senate and House of Representatives.‖).
55. See infra note 78 and accompanying text.
56. See U.S. CONST. art. I, § 7, cl. 1–2 (stating how legislation may be introduced by the House
and the Senate, and laying out the President‘s veto power).
57. See Myers v. United States, 272 U.S. 52, 164 (1926) (holding that the President‘s power to
appoint officers entails the power to remove them, but that the Appointments Clause does not
require the Senate‘s consent to the removal).
58. See U.S. CONST. art. II, § 2, cl. 1 (―The President . . . shall have Power to grant Reprieves
and Pardons for Offences against the United States, except in Cases of Impeachment.‖).
59. Tom Ginsburg, James Melton & Zachary Elkins, On the Evasion of Executive Term Limits,
52 WM. & MARY L. REV. 1807, 1819, 1834–35 (2011) (discussing the development of informal
presidential term limits and the eventual ratification of the Twenty-Second Amendment).
60. See, e.g., U.S. CONST. art. I, § 2, cl. 5 (granting the House the power to impeach executive
officials); id. art. I, § 3, cl. 6 (granting the Senate the power to try all impeachment cases); id. art. II,
§ 2, cl. 2 (limiting the President‘s power to make treaties and appointments to those made ―with the
Advice and Consent of the Senate‖).
61. See Ginsburg, Melton & Elkins, supra note 59, at 1834–35 (explaining that George
Washington‘s service of only two terms led to the creation of an ―unwritten constitutional norm‖).
62. See WILLIAM G. HOWELL, POWER WITHOUT PERSUASION: THE POLITICS OF DIRECT
PRESIDENTIAL ACTION 84 & fig.4.1, app. at 189–91 (2003) (demonstrating the increase in
―significant executive orders‖ during the twentieth century across a diverse set of policy categories).
63. See Jide Nzelibe, A Positive Theory of the War Powers Constitution, 91 IOWA L. REV. 993,
996–97 & nn.2–4 (2006) (discussing the debate over executive powers as one between pro-
President scholars, who stress the importance of strength and flexibility in an executive, and pro-
Congress scholars, who argue that a legislative check on the President‘s foreign-policy power
encourages democratic accountability).
archetypical U.S. model of a constitutionally weak president. It may be that

Latin American presidents look, on paper, very much like their North
American counterparts, in which case we would be tempted to think of
presidentialism in a somewhat more systemic light. On the other hand, it is
very possible that drafters of Latin American constitutions have sculpted a
kind of presidentialism that bears a strong regional cast, which deserves more
systematic description. How presidents stray from their constitutional
prerogatives is, again, a matter that is left open.64
III. Latin American Presidentialism

To what degree can we speak of a Latin American style of
presidentialism? Can we identify features of executive–legislative relations
that are distinctly Latin American or distinctly non-Latin American? How
closely do Latin American constitutions follow the United States archetype,
or for that matter, other relevant models such as the Spanish 1812 (Cádiz)
constitution? Is there, as Loveman claims, a set of provisions unique to the
Latin American constitutions that enables the tyranny that has so frequently
surfaced in these countries?65
Our basic sources are the constitutional documents themselves. We use
the data assembled by the Comparative Constitutions Project (CCP), a com-
prehensive inventory of the provisions of written constitutions for all
independent states between 1789 and 2006.66 Collection of the data is
ongoing, and for purposes of this Article, the dataset includes 647 of the 835
constitutional systems identified by Elkins, Ginsburg, and Melton.67 Elkins,
Ginsburg, and Melton include a large number of questions in their survey
instrument, many of which have to do with the powers of the executive and
the legislature.68 It is this set of questions that constitutes the basic infor-
mation we use here.
The period from independence through the end of the 1870s was one of
intense constitutional experimentation in Latin America. From 1810 through
2007, the nineteen Latin American countries that exist today designed a total
of 231 constitutional systems, 111 of which were written before 1880.
64. We thank John Ferejohn for forcing us to clarify this point.

65. See infra note 80 and accompanying text.
66. COMPARATIVE CONSTITUTIONS PROJECT, http://www.comparativeconstitutionsproject.org.
For details on the conceptualization and measurement of constitutions and constitutional systems,
see Conceptualizing Constitutions, COMPARATIVE CONSTITUTIONS PROJECT,
http://www.comparativeconstitutionsproject.org/conceptualizingconstitutions.htm.
67. A constitutional system consists of a constitution and all its amendments before the
constitution is formally suspended or replaced. We use only one event per system in this analysis—
typically a new constitution in the first year of its adoption.
68. See Zachary Elkins, Tom Ginsburg & James Melton, The Comparative Constitutions
Project: A Cross-National Historical Dataset of Written Constitutions (Survey Instrument),
COMPARATIVE CONSTITUTIONS PROJECT, 18–81 (May 11, 2010),
http://www.comparativeconstitutionsproject.org/files/surveyinstrument.pdf (devoting sixty-four
pages to questions about the powers of the executive and the legislature).
Appendix A provides some sense of the population and our sample. Our
sample includes 193 of the 231 systems, or 81%. The thirty-eight
constitutions missing from our sample tend to be concentrated in the early
years after independence.69 Of the 111 systems adopted (and discarded)
before 1880, our sample includes seventy-nine systems—roughly four-fifths
of that population. These early years are precisely the years of institutional
vacuum that followed independence, when there was the highest degree of
constitutional experimentation. That this population of systems is
underrepresented in our sample implies that our estimate of intraregional
diversity may be biased towards increased homogeneity in the first decades
of the nineteenth century.
We start by considering a set of thirteen attributes, ten pertaining to
powers allocated to the executive and three to powers allocated to the
legislature. Regarding the executive, we consider the following powers: to
issue executive decrees, to assume emergency powers, to propose constitu-
tional amendments, to propose the budget law, to initiate regular legislation,
to veto legislation, to issue pardons, to appoint and dismiss the cabinet, and
to dissolve the legislature. Regarding the legislature, we consider the
legislature‘s power to remove individual ministers, to exercise oversight over
the executive, and to override the executive veto (assuming the constitution
provides for such veto).
Appendix B presents the proportion of Latin American constitutions
with selected executive–legislative provisions over time. For the temporal
dimension, we divide the region‘s history into five eras: (1) a period of
economic and political disorganization (independence through 1870); (2) the
period of agro-export development, during which most countries in the
region were integrated into the international economy as exporters of raw
material and importers of industrialized goods (1870–1918); (3) the period of
crisis of the export model and emergence of import-substitution
industrialization (1919–1945); (4) the period of dominance and then decline
of import-substitution industrialization (1946–1979); and (5) the period of
democratization and economic reforms (1979–2007).
This table displays a remarkable evolution in executive powers across
Latin American constitutions. Let us start, however, with the less remarkable
features of the table. Given that most Latin American constitutions have
been presidential, it is not surprising that the number of constitutions that
allow presidents to dissolve the legislature is relatively small. This number,
however, is not trivial: overall, there have been seventeen Latin American
constitutions that allowed the executive to dissolve the legislature; of these,
eleven are classified as presidential. Equally unsurprising is the fact that
close to 90% of the constitutions written since independence have granted the
69. Of the thirty-eight cases not sampled, thirty-two are constitutions that were written before
1860.
executive the power to freely appoint and dismiss the cabinet. And virtually
every Latin American constitution grants the executive emergency power
(although there is considerable variation regarding the specifics of this
power, as we will see below). Finally, many constitutions have established
relatively strong legislatures, at least when it comes to oversight of the exec-
utive (a feature that has been almost universal since 1870), removal of
individual ministers (about one-half of all Latin American constitutions so
allow), and override of an executive veto (almost all of the post-World
War II constitutions provide for it).
The remarkable development, in our view, is the increase in provisions
that grant the executive some lawmaking powers. A high proportion of
executives have always been given decree powers in Latin American
constitutions, but twentieth-century constitutions rendered this provision
almost universal (although, again, there is considerable variation in the
specifics of this power, as we will see below). Equally prevalent has been
the executive‘s veto power: close to 90% for the whole period and universal
for the post-1979 period. But, whereas less than 10% of the constitutions
written in the nineteenth century allowed the executive to propose constitu-
tional amendments, the proportion in the post-1979 period has soared to
90%. Although less dramatic, a similar pattern is evident with respect to the
executive‘s capacity to initiate ordinary legislation and to propose budget
legislation.
Thus, we see some convergence in Latin American constitutions in the
sense of an expansion of the powers of the executive, particularly executive
lawmaking powers. At the same time, powers that were relatively common
in earlier constitutions either did not change much or have expanded in more
recent times. This pattern can be observed in Appendix C, which plots the
proportion of constitutions in force that provide for a given power. This
convergence includes the features normally associated with presidential
constitutions—the executive‘s power to appoint and dismiss the cabinet, and
the inability to dissolve the legislature. One preliminary observation may be
that the data suggest a contemporary pattern of Latin American constitution-
alism that combines a strong legislature with a president possessing strong
lawmaking powers. This contrasts with the earlier pattern of strong legisla-
tures with presidents possessing few or no lawmaking powers.70
How unique is this pattern with respect to other presidential
constitutions? Is the evolution of Latin American constitutions toward
broader legislative power for the executive a region-specific development, or
70. Interestingly, the earlier pattern has been identified by Shugart and Carey as a configuration
conducive to regime survival, while the current configuration is viewed by them as detrimental to
successful governments. MATTHEW SOBERG SHUGART & JOHN M. CAREY, PRESIDENTS AND
ASSEMBLIES 277 (1992). A test of this proposition is beyond the scope of this Article, but we note
that the earlier period was associated with instability in constitutional form. See supra text
accompanying note 69.
is it part of an overall trend, if not in all constitutions, at least in presidential

constitutions outside Latin America? Appendix C, which also plots the Latin
American trend against the trend in non-Latin American presidential
systems, provides some answers to this question. Note that we plot the non-
Latin American systems starting in 1940; before that time, there are not
enough cases in that subgroup to justify any sort of generalization.71 In
eleven of the thirteen provisions plotted in Appendix C, we observe signifi-
cant separation between the Latin American and non-Latin American
presidential systems (panels 1, 2, 3, 4, 5, 6, 8, 10, 11, 12, and 13). Six of
these eleven differences are in the direction of more executive power for
Latin American presidents compared with non-Latin American ones (panels
2, 5, 10, 11, 12, and 13); two additional differences are in the direction of
more power to the Latin American legislatures over the executive compared
with non-Latin American ones (panels 6 and 8). Only in provisions to dis-
solve the legislature do non-Latin American presidents appear to have an
edge in power (panel 1), although this edge is seemingly disappearing.
Furthermore, four of the five items in which Latin American presidents
exhibit comparatively high power are provisions that can be broadly charac-
terized as lawmaking powers—powers that are, in a sense, shared with the
legislatures (panels 10, 11, 12, and 13).
We can explore these comparisons in more aggregate fashion by
assessing the similarity between any two constitutions across dimensions of
executive–legislative relations. We calculate this quantity simply by compu-
ting the proportion of the thirteen features that we considered in Appendix B
for which any two constitutions agree (in that they both either include or
exclude the provision in their constitutions). Appendix D describes the mean
of these measures across various subgroups (all presidential constitutions,
Latin American presidential constitutions, non-Latin American presidential
constitutions, and non-Latin American, non-presidential constitutions). On
average, any two constitutions in the data share nine of thirteen provisions
for a score of 0.68. The first thing to note is that presidential systems in gen-
eral are a more coherent category than either parliamentary or semi-
presidential systems. Latin America accounts for the vast majority of presi-
dential systems before 1945; indeed, it is possible that the overall coherence
of the presidential category is driven by the similarity of constitutions within
the region. After 1945, there is increasing divergence between Latin
American and other presidential systems. Non-Latin American presidential
71. Before 1940, there were nine presidential constitutions outside of Latin America: the United
States (1789), Haiti (1843 and 1935), France (1848), Germany (1919), Lithuania (1938), Liberia
(1847), and the Philippines (1899 and 1935). Since 1940, there have been eighty presidential
constitutions written in countries outside of Latin America. Note that the 1919 German constitution
did not explicitly provide for a directly elected president, and for this reason it is not classified as a
semi-presidential constitution. For the classification of constitutions as presidential, parliamentary,
and semi-presidential, see generally Cheibub, Elkins & Ginsburg, supra note 9.
systems exhibit the same level of coherence as non-presidential systems after

1945, while presidential systems within the region seem to be becoming
more similar.
The similarity between Latin American constitutions and the United
States Constitution is not particularly high, relative to other models. The two
other models that influenced Latin American constitution makers after inde-
pendence were France‘s constitution of 1791 and Spain‘s 1812 (Cádiz)
constitution.72 Both of these constitutions (as well as their close cousins, the
Portuguese constitution of 1822 and the Norwegian constitution of 1814)
carved out a subordinated position for the monarch in an otherwise-
republican document and represented the leading alternative model.73 Latin
American constitutions are not especially similar to any one of these
documents. The mean similarity between Latin American constitutions
across the sample and each of these documents is not significantly different
from the mean similarity of any two constitutions. Thus, even though Latin
Americans ultimately settled on the form of government conceived by their
North American brethren—a president popularly elected for a fixed term in
office—they did not necessarily adopt the same ancillary provisions regard-
ing the specific allocation of powers between the executive and the
legislature. This suggests that the adoption of presidentialism in Latin
America was less the product of automatic or mechanistic borrowing from
the U.S. Constitution and more the adoption of a particular institutional solu-
tion discovered by the North Americans to the problem that Latin Americans
were facing: how to establish a national executive once the monarch had
been removed. Nevertheless, the executive that they designed had as much
in common with the Spanish Prime Minister as it did with the U.S. President.
IV. Executive Lawmaking Power as a Signature Feature of Latin American

Presidentialism
To the extent that Latin American constitutions represent a distinct
breed of presidentialism, the distinction is manifested in the strong
lawmaking power that they vest in the president. By lawmaking power, we
mean here the powers of emergency, decree, and the initiation of constitu-
tional amendment and legislation. We examine these provisions in some
detail below.
72. See LOVEMAN, supra note 46, at 54 (acknowledging newly formed Latin American nations‘
incorporation of rights and liberties from the Cádiz constitution and French Revolutionary ideals).
73. See JOHN A. HAWGOOD, MODERN CONSTITUTIONS SINCE 1787, at 49–58 (Fred B.
Rothman & Co. 1987) (1939) (comparing the Portuguese, French, Spanish, and Norwegian
constitutions of the era, and noting the limits on the monarch‘s powers in each); LOVEMAN, supra
note 46, at 40–45 (describing the limited role of the Spanish monarch under the Cádiz constitution).
A. Emergency Powers
A word is in order as to why we consider emergency powers to be
legislative in nature. First, periods of emergency rule generally allow for the
temporary delegation of considerable powers—including those normally
vested in the legislature—to the executive.74 The easier it is to declare a state
of emergency, the more likely it will be that the executive will predominate
and in some cases even usurp legislative authority strategically. Second, the
executive may be able to act without legislative authorization, as Ferejohn
and Pasquino recognized in their study distinguishing between constitutional
and legislative models of emergency powers.75 In their legislative model,
ordinary legislation facilitates emergency power, and so there is not a true
―regime of exception‖76 outside constitutional constraints.77 But much
depends on the specific assignment of powers to declare an emergency and
then to legislate during one.
The U.S. Constitution provides for relatively narrow emergency powers.
The relevant clause provides that ―[t]he Privilege of the Writ of Habeas
Corpus shall not be suspended, unless when in Cases of Rebellion or
Invasion the public Safety may require it.‖78 By contrast, Bolivia‘s 1851
emergency provision imagines broader powers for the president (―to assume
extraordinary faculties‖) but also requires the consent through countersigna-
ture of all ministers of state in order to establish emergency conditions.79
This is a model with a legal constraint, but it does not fit the legislative
model fully because the legislature has no involvement. According to
Loveman, it is precisely these sorts of expansive emergency provisions that
74. See John Ferejohn & Pasquale Pasquino, The Law of the Exception: A Typology of
Emergency Powers, 2 INT‘L J. CONST. L. 210, 217 (2004) (―The legislative model handles
emergencies by enacting ordinary statutes that delegate special and temporary powers to the
executive.‖); Mark Tushnet, The Political Constitution of Emergency Powers: Parliamentary and
Separation-of-Powers Regulation, 3 INT‘L J.L. CONTEXT 275, 275 (2007) (―‗Emergency powers‘
describes the expansion of governmental authority generally . . . , and the transfer of important ‗first
instance‘ lawmaking authority from legislatures to executive officials, in emergencies.‖).
75. Ferejohn & Pasquino, supra note 74, at 211–21.
76. See LOVEMAN, supra note 46, at 6 (establishing that many Latin American constitutions
contained provisions allowing the invocation of ―regimes of exception,‖ wherein executive
authority would be expanded, and constitutional protections, rights, and liberties would be
temporarily voided).
77. See Ferejohn & Pasquino, supra note 74, at 219 (―[B]ecause the legislature—the part of the
government closest to the people—actively delegates authority to the executive, the exercise of that
power is more constrained and legitimate and is even, indeed, amplified and made more efficient by
the fact that this exercise is supported by the legislature and, presumably, by the people.‖).
78. U.S. CONST. art. I, § 9, cl. 2.
79. CONSTITUCIÓN POLÍTICA DE LA REPÚBLICA DE BOLIVIA [BOL. CONST. 1851] Sept. 21,
1851, art. 76, cl. 26.
have threatened the political stability in the region almost from the
beginning.80
Nevertheless, emergency provisions vary in important ways across
Latin American constitutions. Appendix E, which reports the proportion of
constitutions that contain various emergency provisions, demonstrates some
of this variation. The variation spans at least three dimensions: (1) the iden-
tity of the actors (in addition to the executive) involved in the process of
declaring the existence of an emergency situation; (2) the conditions under
which emergency can be declared; and (3) limitations on the actions taken
under emergency conditions. The last two rows of Appendix E present,
respectively, the proportion of constitutions that require the legislature to
play some role in the process of declaring an emergency (i.e., it must
approve, or at least be consulted before, the declaration of an emergency),
and the proportion of constitutions that explicitly specify the conditions
under which an emergency may be declared. Across these dimensions, we
focus on four specific aspects of emergency provisions: (1) the participation
of the legislature in the emergency process; (2) the reference to internal secu-
rity reasons as a justification for emergency powers; (3) the explicit provision
for the suspension or restriction of rights during emergency; and (4) the pro-
hibition of constitutional amendments during emergency rule.
Several patterns are worth noting. The first is that almost all
presidential constitutions contain emergency provisions, whereas 73.6% of
parliamentary and 81.0% of semi-presidential constitutions do. Indeed,
every Latin American presidential constitution written since independence
contains a provision for the executive to declare an emergency, compared
with 94.4% of non-Latin American presidential constitutions.
The role of the legislature in the process of emergency declaration is
smaller in presidential constitutions than it is in either parliamentary or semi-
presidential ones. Only 19.0% of Latin American presidential constitutions
require that the legislature approve the state of emergency, and an additional
1.9% require that the legislature at least be consulted, whereas 43.5% of non-
Latin American presidential constitutions require some form of legislative
participation (approval or consultation). There is a trend toward increasing
the participation of legislatures in declaring emergencies, but this trend is
weaker in presidential constitutions in Latin America than elsewhere: of the
more recent (post-1979) constitutions, 36.8% of Latin American presidential
constitutions require legislative participation, while 55.9% of non-Latin
American, non-presidential constitutions do.
It is likely that the criteria stipulated in a constitution for identifying an
emergency tell us something about the flexibility of the executive‘s power
under these conditions. We can speculate that constitutions that limit
80. See LOVEMAN, supra note 46, at 6–9 (stating that while the clauses granting emergency
powers ―did not cause violence and dictatorship,‖ they ―are the constitutional foundation for tyranny
almost everywhere in Latin America today‖).
emergencies to natural disasters are less flexible than ones that allow for
emergencies for public security reasons. Further, we can speculate that con-
stitutions that allow for emergencies in response to internal security issues
are particularly broad, since they do not require an external trigger such as an
invasion. We find that reference to internal security reasons as a justification
for the state of emergency is more common in presidential than it is in par-
liamentary or semi-presidential constitutions; it is more common in Latin
American than in non-Latin American presidential constitutions; and
although it has become more common in all constitutions over the years, it is
considerably more common in Latin American presidential than in non-Latin
American, non-presidential constitutions. Emergency rule seems easier to
invoke in Latin America than it does elsewhere.
Presidential and parliamentary constitutions are equally likely to contain
an explicit provision allowing for the suspension or restriction of rights dur-
ing emergency rule. Among presidential constitutions, however, those in
Latin America are considerably more likely to allow for the suspension of
rights than those outside of Latin America. The proportion of Latin
American presidential constitutions with such a provision has hovered
around 90% in the post-World War II period.
Finally, only a small proportion of all constitutions explicitly forbid
legislative dissolution or constitutional amendments during emergency rule.
It is not surprising that this proportion is much smaller in presidential than in
parliamentary and semi-presidential constitutions, since assembly dissolution
by the executive is not a common item on the ordinary menu of presidential
powers outside assembly-confidence systems.81 Latin American presidential
constitutions, however, are about one-fifth as likely as non-Latin American
presidential constitutions to contain a provision prohibiting legislative
dissolution under emergency rule. We do not know if this distinction reflects
the fact that a Latin American presidency is particularly empowered vis-à-vis
the legislature (because it is free to dissolve the assembly) or disempowered
(because it is never allowed to dissolve the assembly and so the constitution
is silent about the rule during emergencies), but it is at least possible that the
former is the case. To summarize, Latin American presidential constitutions
are relatively less likely to require some form of legislative participation for
the activation of emergency powers; more likely to permit internal security
concerns as justifying the state of emergency; and more likely to explicitly
allow for the suspension or restriction of rights. This is largely consistent
with Loveman‘s claims.82 It seems unlikely, however, that the presence of
these provisions exhibits any causal relationship with the instability that has
81. See CHEIBUB, supra note 13, at 10 (remarking that the threat of dissolution is ―absent, by
design,‖ from presidential constitutions); cf. Ginsburg, Melton & Elkins, supra note 59, at 1816
(noting that in popular-election systems, the legislature and executive are constituted independently,
while in assembly-confidence systems, either branch can dissolve the other).
82. See supra note 80 and accompanying text.
characterized Latin American republics since independence, as Loveman

suggests. Some of the features that might have been conducive to providing
incentives for executive unilateral action—such as the lack of legislative
involvement, the suspension of rights, and the reference to internal
security—are relatively new developments; the early constitutions, which
according to Loveman provided the foundations of tyranny in the region, did
not possess all of these provisions.
B. Decree Powers
Executive decree powers give the executive the ability to issue binding
rules with the force of law and are an important feature of modern
governments,83 being found in about two-thirds of all constitutions. The
design of decree powers varies widely across cases, as illustrated in
Appendix F, and their rationale and distribution depends on the broader
political system: about 70% of both semi-presidential and presidential con-
stitutions provide such power for the executive, while only half of
parliamentary systems do so. In systems with fused governmental powers
(parliamentary and some semi-presidential constitutions), the decree power
for the executive is usually conceived as the exercise of delegated power
from the legislature. The legislature, therefore, is frequently designated as
the body that must approve an executive decree in those systems. In contrast,
only 27.0% of presidential constitutions designate the legislature as the
approving body of executive decree powers. Instead, such systems often
require that the executive approve the decree, meaning in practical terms the
cabinet in most cases. This is consistent with a conception of separation of
powers and the notion of a discrete realm of executive lawmaking.
In neither system is it the case that executives are unconstrained in their
ability to issue decrees. In fact, the difference between the three systems
almost disappears when we consider whether the constitution specifies that
some governmental body—be it the legislature or the cabinet—must approve
executive decrees. The numbers (not shown in the Appendix) are 68.3% for
presidential constitutions and 73.2% for both parliamentary and semi-
presidential constitutions.
In keeping with the logic of fused powers and delegated authority,
parliamentary and semi-presidential constitutions are twice as likely as
presidential constitutions to stipulate that, once issued, executive decrees are
immediately effective. Presidential constitutions are significantly more
likely than parliamentary and semi-presidential constitutions to require that
the approving body (the legislature or the cabinet or both) approve the decree
before it becomes effective. Thus, at least in this respect, the executive is
83. See John M. Carey & Matthew Soberg Shugart, Calling Out the Tanks or Filling Out the
Forms?, in EXECUTIVE DECREE AUTHORITY 1, 9, 15–19 (John M. Carey & Matthew Soberg
Shugart eds., 1998) (defining decree as ―the authority of the executive to establish law in lieu of
action by the assembly,‖ and discussing its appeal as a component of democratic government).
more constrained in presidential than in parliamentary and semi-presidential

constitutions.
When it comes to the validity of the decree, there are two basic
situations. In the first, the decree, once issued, is permanent unless it is
explicitly rejected by the legislature; in the second, the decree expires after
its pre-specified duration period unless it is explicitly extended by the
legislature. The first situation favors the executive: the decree becomes the
status quo and the legislature must act in order to change it. The second situ-
ation favors the legislature: the status quo ante is restored unless the
legislature prefers the situation generated by the decree. The biggest differ-
ence we observe across systems is that presidential constitutions are less
likely to specify who must act, and in what way, once the executive decree is
issued: only 18.4% of presidential constitutions (as compared with 36.5% of
parliamentary and 28.3% of semi-presidential) clearly state what must ensue
after the decree is issued.84 This, of course, allows for a degree of ambiguity,
the result of which cannot be specified in the abstract. Our guess is that the
lack of specification is a problem for the working of these constitutions.
As with emergency powers, some features of executive decree
regulation seem to characterize a particularly Latin American model of
presidentialism. First, constitutions with executive decree power are more
frequent in Latin American presidential constitutions than in non-Latin
American presidential constitutions. While almost half of non-Latin
American presidential constitutions render the executive decree immediately
valid upon issuance, only a small fraction (5.7%) of Latin American presi-
dential constitutions do the same. However, when not left unspecified (as
42.7% of Latin American constitutions do), 46.8% of Latin American presi-
dential constitutions require the action of the approving body before the
decree becomes effective. This implies a routinization of presidential decree
making, though we do not know the extent to which these formal constraints
actually serve to prevent presidents from pursuing their preferred policies.
Finally, Latin American presidential constitutions are far more likely (87.9%,
as compared to 44.0% for non-Latin American presidential constitutions) to
leave the issue of decree validity unspecified, remaining silent about what
happens once the decree is issued. To the extent that the decree changes the
status quo and the constitution is silent as to whether the status quo ante can
be restored, we believe that this lack of regulation tends to favor the
executive—though it is hard to be sure in the absence of more detailed
information on de facto practices.
84. These numbers refer to the sum of the rows labeled ―Permanent, unless repealed‖ and
―Naturally expires, unless extended‖ in Appendix F.
C. Constitutional Amendment and Initiation of Legislation

It is not surprising that most constitutions provide some mechanism for
their amendment.85 But not all of them allow the executive to propose such
amendments. As we can see in Appendix G, only 43.2% of the world‘s con-
stitutions that specify amendment procedures allow the executive to propose
constitutional amendments. This proportion is considerably higher in semi-
presidential than in parliamentary and presidential constitutions; among
presidential constitutions it is higher in non-Latin American constitutions,
although almost 90% of the more recent post-1979 Latin American presiden-
tial constitutions allow the president to propose constitutional amendments.
In addition to constitutional amendments, a large proportion of
constitutions provide for a legislative process involving budget bills. Again,
this is not surprising since the budget is probably the most important piece of
legislation that comes regularly before a legislative body. It is interesting to
observe, however, that even though a large proportion of presidential,
parliamentary, and semi-presidential constitutions provide for an explicit
legislative process around the budget bill, almost half of presidential consti-
tutions allow the executive to initiate the budget bill, compared to less than
one-fifth of parliamentary and semi-presidential constitutions. Moreover,
Latin American presidential constitutions are almost twice as likely as non-
Latin American presidential constitutions to allow the executive to initiate
budget bills. This is true for all historical periods and has increased in the
more recent periods.
It is commonly argued that presidential constitutions do not provide a
constitutional mechanism to break deadlocks or impasses between the legis-
lature and the executive when they emerge.86 The fixed nature of the
legislative and executive terms, it is argued, deprives political actors of the
opportunity to remove the government constitutionally when a crisis
emerges.87 Yet, as Appendix G indicates, at least when it comes to the
budget, a large proportion of constitutions stipulate what should happen in
case a budget is not approved. Whereas it is true that presidential
constitutions—as compared to parliamentary and semi-presidential—are
least likely to specify the default situation in case the budget bill fails, over
half of these constitutions still do so. In presidential constitutions, the prac-
tice is to either adopt the previous year‘s budget or to adopt the budget that
85. In our sample, there are nine out of 444 (representing 1.99%) constitutions that do not
explicitly provide for a revision mechanism; two are presidential, six are parliamentary, and one is
semi-presidential. None of these is in Latin America.
86. See, e.g., Shugart & Mainwaring, Rethinking the Terms of the Debate, supra note 12, at 32
(observing that constitutional mechanisms for resolving these kinds of conflicts are of ―doubtful
democratic legitimacy‖).
87. See, e.g., id. at 30 (explaining that while most presidential systems with fixed executive
terms have provisions for impeachment, ―they offer less flexibility in crisis situations because
attempts to depose the president can easily endanger the regime itself‖).
was proposed by the executive. Other solutions, including adopting the

budget proposed by the legislature, are less commonly adopted.
As to the other types of laws—organic laws, finance, tax, and spending
bills—a considerably smaller proportion of constitutions specify a legislative
process to approve them, and among those that do so, the proportion that
allows for the executive to initiate them is also relatively small, with the
exception of spending bills. There is no discernible pattern across regime
type and region when these processes are considered together. The only
noticeable thing is that post-World War II constitutions are more likely to
specify legislative processes around these various bills, and when they do so,
they are more likely to allow the executive to initiate them.
Thus, the trend we identified earlier regarding increasing powers of
legislative initiative granted to presidents in Latin America is primarily due
to the fact that, in this region, presidents are allowed to set the agenda when
it comes to constitutional amendments and budget laws. These are probably
the two most important regular legislative activities in any political system,
and granting the executive such powers is of great significance in terms of
overall political impact. In short, the executive is a legislative leader in Latin
America.
V. Is Presidential Lawmaking Desirable?

Our analysis has emphasized the concentration of lawmaking authority
in the executive, a trend that has occurred over time in many political
systems, but one that we have argued has been especially pronounced in
Latin America. This is of course a major departure from the Montesquieuan
conception of separated powers, in which lawmaking is done by the legisla-
ture and the only role of the executive is to execute the laws.88 Such a
conception was highly influential for the American founders, whose design
of a constitutional scheme shaped the approach of subsequent constitution
makers.89 In the eighteenth century, the separation of powers scheme was
88. See M. DE SECONDAT, BARON DE MONTESQUIEU, THE SPIRIT OF THE LAWS 162–63, 172
(J.V. Pritchard ed., Thomas Nugent trans., Fred B. Rothman & Co. 1991) (1914) (noting that the
legislative power enacts the laws and the executive carries out all functions of the state not reserved
to the judiciary, but that the executive ―has no other part in the legislative [power] than the privilege
of rejecting‖).
89. See Carl T. Bogus, The Battle for Separation of Powers in Rhode Island, 56 ADMIN. L.
REV. 77, 91 (2004) (―Montesquieu especially influenced the American Founders on the concept of
separation of powers.‖); Susanna Frederick Fischer, Playing Poohsticks with the British
Constitution? The Blair Government’s Proposal to Abolish the Lord Chancellor, 24 PENN ST. INT'L
L. REV. 257, 283 (2005) (―Montesquieu‘s views on the separation of powers are at least somewhat
familiar to most Americans, because his writings had such a profound influence on some of the
American Founders.‖); Ken I. Kersch, Justice Breyer’s Mandarin Liberty, 73 U. CHI. L. REV. 759,
780 (2006) (book review) (―[T]he Constitution, . . . and the American people, were fully committed
to government by elected representatives, an independent judiciary, [and] separation of powers
more generally . . . , thanks in large part to the influential writings of [the] liberal French thinker,
Montesquieu.‖). For a discussion of the influence of the American founders on other constitutional
drafters, see Bruce Ackerman, The New Separation of Powers, 113 HARV. L. REV. 633, 634–42
seen as normatively attractive to prevent tyranny.90 We thus must ask

whether the erosion of this separation, particularly through the agglomeration
of lawmaking power in the executive, enables tyranny. We also should try to
understand why the agglomeration has occurred, so as to consider whether
there are offsetting normative advantages.
In understanding the positive question of why the concentration of
power has occurred, one can distinguish two broad positions that are not
completely incompatible. One argument is that the concentration of
lawmaking authority in the executive is a response to the exigencies of
modern government. The other is that the concentration reflects a self-
conscious power grab by the executive. These positions have very different
normative implications: if we think concentration of lawmaking authority is
functional in some sense, then it is obviously more attractive than if it merely
reflects the self-aggrandizement of one branch.
Consider first the functional argument. With the rise of the
administrative state, the scope of government activity has dramatically
increased, with a need for regulation that can respond to changing conditions
in technically complex areas. The executive is the head of the
administration, which is staffed with bureaucrats who have the relevant
policy expertise to make such decisions. Thus, the apparent concentration of
lawmaking authority in the executive hides a dispersion of power within the
larger administrative state. But the fact of administrative lawmaking is a
necessary response to complexity.
This argument helps one to understand why we would observe the
expansion of legislative initiative within the executive branch. Experts who
are charged with solving problems and adjusting regulations to changing cir-
cumstances may want to be proactive in lawmaking and not simply wait for
the generalist legislature to take the lead. A presidential initiative is an
acknowledgement of the fact that it is the executive that will make the rele-
vant decisions about the content of regulation.
Similarly, the expansion of executive decree power may in part reflect
the need for technical regulation that every political system faces. Whether
under delegated authority from the legislature or under powers assigned
directly to the executive, the modern administrative state requires that the
technical details of complex regulatory schemes be made by experts. Decree
authority is one mode of such lawmaking.
The concentration of lawmaking in the presidency in particular provides
for another functional advantage: accountability. In the United States, it has
(2000); George Athan Billias, Introduction to AMERICAN CONSTITUTIONALISM ABROAD, supra

note 16, at 1, 1–6.
90. See Douglas W. Kmiec, Debating Separation of Powers, 53 REV. POL., 391, 393 (1991)
(book review) (describing how the delegates of the Constitutional Convention of 1787 relied on
Montesquieu‘s notion of separation of powers to ―devise a check upon legislative dominance that
would not itself devolve into tyranny‖).
been argued that the rise of the so-called ―plebiscitary presidency‖ has
changed the structure of the office.91 Presidents are typically the only figures
elected by a national constituency and hence are more likely to reflect the
preferences of the median voter.92 Congress, in contrast, is seen as respond-
ing to a myriad of local interests, and hence it is not expected to produce
policies truly in the national interest.93 Furthermore, policy in Congress is
produced through a complex process of committees, vote trading, and nego-
tiation across houses, which makes it difficult to assign responsibility for any
particular policy. When a single individual holds responsibility, the public
clearly knows whom to blame or credit for policies. Executive lawmaking,
in this view, facilitates accountability.
In contrast with these functional accounts, some have asserted that the
concentration of authority in the presidency reflects a naked power grab.
This is the view associated with Loveman and others who argue for the con-
tinuing relevance of the caudillo tradition in Latin America.94 These scholars
emphasize the use of the emergency power by Latin American presidents.95
The emergency power, they show, has long been used to take power from the
legislature and leads to periods of executive tyranny.96 The assignment of
decree power to the executive, in this view, also comes at the expense of the
legislature, in that the executive can use that power not only with regard to
the technical details of delegated lawmaking, but also for setting the broad
outlines of policy.97
A full evaluation of these competing positions is beyond the scope of
this Article, but we lean toward the view that there is something quite func-
tional about the expansion of executive lawmaking authority. There are two
reasons for our view. First, we observe the increasing power of single
91. See Jide Nzelibe, The Fable of the Nationalist President and the Parochial Congress, 53
UCLA L. REV. 1217, 1224–31 (2006) (contrasting the modern plebiscitary vision of the presidency
with the early Federalist vision, and surveying the modern scholarship and judicial conclusions
regarding the structural implications of a plebiscitary presidency).
92. See INS v. Chadha, 462 U.S. 919, 948 (1983) (observing that the President brings a
―national‖ perspective to the legislative process); Elena Kagan, Presidential Administration, 114
HARV. L. REV. 2245, 2335 (2001) (―[B]ecause the President has a national constituency, he is likely
to consider, in setting the direction of administrative policy on an ongoing basis, the preferences of
the general public, rather than merely parochial interests.‖).
93. Chadha, 462 U.S. at 948 (citing Myers v. United States, 272 U.S. 52, 123 (1926)).
94. LOVEMAN, supra note 46, at 398; see also R.A. HUMPHREYS, TRADITION AND REVOLT IN
LATIN AMERICA AND OTHER ESSAYS 220 (1969) (―[T]he caudillo tradition survives. Political
creeds exist, and some of them are increasingly important.‖).
95. See, e.g., LOVEMAN, supra note 46, at 5–6 (―Latin American constitutions almost always
included provisions for ‗emergency powers,‘ or ‗extraordinary powers‘ . . . [that] might be exercised
by presidents . . . .‖).
96. See, e.g., id. at 6 (―[N]ormal constitutional protections were suspended, rights and liberties
were temporarily voided, and the government‘s authority was greatly expanded.‖).
97. See id. at 21 (―[O]rdinary government procedures for legislation, administration, and
judicial decision making may be replaced temporarily with special methods for making and
implementing public policy.‖).
individuals even in pure parliamentary systems, where scholars have spoken

of the presidentialization of the office of prime minister—a phenomenon that
has resulted from the structure of political parties and the ubiquity of media
coverage of politics.98 This suggests that there is indeed something to the
argument that having a single individual at the center of the political system
enhances accountability. Second, the argument about tyranny is largely
rooted in historical experience rather than contemporary reality. It assumes
that long-run institutional patterns of behavior are enduring. For much of
Latin American history, this was an understandable position. But we are
now in an era of widespread democratic government in Latin America, with
all the countries of the region observing formal norms of democracy. The
trend toward democracy has accelerated since the 1980s, which covers part
of the period in which we find enhanced powers of executive lawmaking.
We do not assert that the two phenomena are causally related, but their con-
temporaneous occurrence suggests prima facie that democracy is not
incompatible with expanded executive lawmaking. One can contrast the
presidential systems in Africa, which form the bulk of our comparison group
and in which democratic norms are much less frequently observed. As a
normative matter, then, we believe the Latin American presidential pattern is
one to be celebrated rather than condemned.
VI. Conclusion
We have analyzed the formal features of executive power in Latin
America, a region long understood to be one amenable to strong executive
rule. We have demonstrated that, although the presidency was inspired by
the American model, other models were equally influential in structuring the
precise contours of executive and legislative power in the region. We have
also seen increasing convergence within the region along important dimen-
sions of executive–legislative relations. We can thus speak of a Latin
American model of presidential power that includes a powerful role in legis-
lation as well as extensive emergency rule. This distinguishes the Latin
American presidency from those in other regions of the world.
Our analysis has several implications for the study of comparative law
and politics. First, it calls attention to geography as an important predictor of
constitutional design. Second, our analysis emphasizes change rather than
continuity and convergence over time. This approach contrasts with the
recent emphasis in comparative law on ―legal origins‖ as determinants of
98. See Thomas Poguntke & Paul Webb, The Presidentialization of Politics in Democratic
Societies: A Framework for Analysis, in THE PRESIDENTIALIZATION OF POLITICS 1, 5–6 (Thomas
Poguntke & Paul Webb eds., 2005) (explaining that the degree to which presidentialization occurs
in any system, including parliamentary ones, depends on a range of factors including ―changes in
the social structure and the media system‖).
contemporary outcomes.99 Finally, while the legal-origins analysts

emphasize the importance of French law in Latin America,100 our account
shows that at a constitutional level, the influence of Spain and the United
States was also significant in the early years. But while the legal-origins
school argues for long-range consequences of initial choices, we observe a
gradual process of constitutional updating in which constitutions within the
region grow more similar to each other, and a move away from the models
from which they were initially drawn.
99. Cf. generally Rafael La Porta et al., Law and Finance, 106 J. POL. ECON. 1113 (1998)
(examining the origins of legal rules covering the protection of corporate shareholders and creditors
in forty-nine countries, as well as the quality of their enforcement).
100. See id. at 1118 (―When the Spanish and Portuguese empires in Latin America dissolved in
the nineteenth century, it was mainly the French civil law that the lawmakers of the new nations
looked to for inspiration.‖).
26
Appendix A. Available and Missing Latin American Constitutions by Decade

Country 1810s 1820s 1830s 1840s 1850s 1860s 1870s 1880s 1890s 1900s 1910s 1920s 1930s 1940s 1950s 1960s 1970s 1980s 1990s 2000s Total
Argentina (1) 1 1 3
Bolivia 1 3 (1) 1 1 2 2 1 1 2 2 17
Brazil 1 1 2 1 1 1 7
Chile 3 1 1 1 6
Colombia (1) 2 1 2 1 1 1 9
Costa Rica (1) 3 (1) 1 1 1 1 1 10
Cuba (1) 1 (1) 1 1 (1) 1 7
Dominican Republic (1) (1) 3 2 (1) 6 2 (1) 1 2 4 1 2 1 4 1 1 34
Ecuador 2 1 (1) 2 2 1 1 1 1 1 2 1 1 1 4 22
El Salvador (1) 1 (1) (1) 1 2 3 1 1 1 13
Gran Colombia (1) 1
Guatemala (1) (1) 1 1 1 1 1 1 8
Honduras (1) (1) (2) 1 1 1 1 1 1 (1) 1 1 1 1 15
Mexico (1) 2 1 1 1 1 7
Texas Law Review
Nicaragua (1) (1) (1) 2 1 1 1 1 (1) 1 1 1 1 14

Panama 1 2 1 4
Paraguay 1 1 1 1 1 1 6
Peru (1) 2 (2) 2 1 2 1 1 1 1 14
Venezuela (2) (1) 1 2 1 1 1 2 3 2 4 2 1 1 1 1 26
United Provinces of
(1) (1) 2
Central America
Uruguay (1) 1 1 1 1 1 6
Total 7 22 17 17 18 14 16 11 7 10 6 13 14 15 9 13 5 7 9 1 231
Note: Numbers in (parentheses) represent missing constitutions.
[Vol. 89:XXXX
Appendix B. Percentage of Latin American Constitutions with

Executive–Legislative Provisions by Year of Promulgation
Pre- 1870− 1919− 1946− Post-
Overall 1870 1918 1945 1979 1979
Executive can dissolve the legislature 9.5 6.4 6.2 17.2 6.3 20.0
Executive can freely dismiss the cabinet 89.5 82.5 91.3 96.6 90.6 95.0
Executive can freely appoint the cabinet 95.3 95.2 100.0 96.6 93.8 85.0
Executive can issue pardons 72.1 63.5 78.3 82.8 78.1 60.0
Executive has veto power 89.5 92.1 82.6 79.3 96.9 100.0
Legislature has oversight powers over the executive 90.0 77.8 95.7 93.1 96.9 100.0
Legislature can remove individual ministers 53.7 61.9 34.8 55.2 62.5 55.0
Legislature can override executive veto 85.8 88.9 78.3 75.9 93.8 95.0
Executive has emergency power 99.0 98.4 100.0 96.6 100.0 100.0
Executive can initiate constitutional amendments 23.7 7.9 4.4 27.6 37.5 90.0
Executive can initiate ordinary legislation 69.0 52.4 65.2 69.0 87.5 100.0
Executive can propose budget 42.1 19.1 21.7 51.7 81.3 85.0
Executive has decree powers 77.3 63.5 71.7 100.0 81.3 95.0
Number of constitutions: 190 63 46 29 32 20

Appendix C. Trends of Executive–Legislative Provisions in Latin

American and Non-Latin American Presidential Systems
Solid line = Latin American constitutions

Dashed line = non-Latin American constitutions
Appendix D. Similarity Between Constitutions with Respect to

Executive–Legislative Features
Presidential Non-Presidential
Latin Non-Latin Same Different
Era All American American system* system*
Pre-1870 0.74 0.75 0.70

990 820 6
1870−1918 0.74 0.75 0.43

990 946 45
1919−1945 0.69 0.73 0.61 0.63 0.61

528 378 10 106 744
1946−1979 0.66 0.78 0.65 0.63 0.58

2,145 406 666 2,965 7,770
Post-1979 0.66 0.82 0.64 0.66 0.61

1,431 120 703 2,088 6,491
All 0.68 0.73 0.63 0.64 0.59

29,403 12,403 3,570 10,626 59,652
Universe: Constitutional dyads (1789–2007). Cells represent the mean proportion of
features that match between two constitutions (above) and the number of dyads
(below).
*―System‖ refers to the classification of constitutions as presidential, parliamentary, or
semi-presidential.
2011]
Appendix E. Regulation of Emergency Provisions in Constitutions that Grant Emergency Powers to the Executive, 1789−2007
Presidential Latin American Presidential Non-Latin American, Non-Presidential
Semi- Pre- 1870− 1919− 1946− Post- Pre- 1870− 1919− 1946− Post-
All Presidential Parliamentary presidential LA Non-LA 1870 1918 1945 1979 1979 1870 1918 1945 1979 1979
Constitutions with emergency provisions 87.2 98.0 73.6 81.0 100.0 94.4 100.0 100.0 100.0 100.0 100.0 91.7 90.0 65.0 75.3 83.8
Body that approves emergency declaration:

Legislature must approve 27.4 25.1 33.7 39.7 19.0 36.5 7.3 18.2 15.2 25.9 31.6 9.1 11.1 24.2 25.0 44.1
Constitutional council must approve 14.5 18.9 12.4 8.8 22.8 11.8 31.7 18.2 23.2 18.5 26.3 6.1 11.1 16.1 10.3 8.6
Legislature must be consulted 6.3 4.1 6.7 22.1 1.9 8.2 2.4 2.3 1.8 0.0 5.3 3.0 0.0 1.6 10.3 12.9
Emergency can be declared:

In case of external war/aggression 52.4 63.4 48.3 54.4 67.1 56.5 51.2 75.0 65.2 63.0 84.2 18.2 38.9 33.9 40.5 53.8
For internal security reasons 44.0 58.4 28.1 39.7 64.6 47.1 58.5 72.7 66.1 48.2 79.0 24.2 33.3 33.9 25.0 37.6
In case of natural disaster 16.8 21.0 23.6 19.1 22.2 18.8 2.4 4.6 8.0 37.0 84.2 0.0 11.1 6.5 6.9 30.1
In situations of general danger 27.2 37.9 30.3 10.3 42.4 29.4 39.0 29.6 36.6 51.6 63.2 6.1 11.1 17.7 19.8 19.4
Economic emergency 5.0 5.4 10.1 0.0 5.7 4.7 0.0 0.0 0.9 14.8 21.1 0.0 0.0 1.6 6.9 5.4
Left to nonconstitutional law 1.8 0.8 3.4 4.4 0.0 2.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 2.6 4.3
Not specified 12.9 7.4 23.6 16.2 4.4 12.9 2.4 6.8 5.4 3.7 0.0 0.0 5.6 11.3 2.6 18.3
Restrictions on emergency powers:

Rights can be suspended or restricted 49.6 60.1 58.4 45.6 69.0 43.5 29.3 75.0 59.8 88.9 94.7 15.2 38.9 38.7 37.1 57.0
Latin American Presidentialism
Legislature cannot be dissolved 6.6 4.5 10.1 19.1 1.9 9.4 2.4 0.0 1.8 3.7 0.0 0.0 0.0 0.0 12.1 11.8
Constitution cannot be amended 1.8 1.2 2.3 7.4 1.3 1.2 0.0 0.0 0.0 3.7 5.3 0.0 0.0 0.0 2.6 4.3
No restrictions are imposed 4.5 5.8 3.4 1.5 5.7 5.9 7.3 6.8 6.3 0.0 10.5 3.0 5.6 1.6 6.9 1.1
Legislature plays a role 32.6 28.8 38.2 58.8 20.9 43.5 9.8 20.5 22.2 25.9 36.8 12.1 11.1 38.5 31.9 55.9
Specification of conditions 65.7 77.8 62.9 66.2 82.9 68.2 75.6 84.1 81.5 85.2 94.7 27.3 50.0 50.0 54.3 66.7
Note: Entries represent the percentage of constitutions in each category (column) that provide for a given feature (row).
31
32
Appendix F. Regulation of Decree Powers in Constitutions that Grant Decree Powers to the Executive, 1789−2007
Presidential Latin American Presidential Non-LA Pres.
Semi- Pre- 1870− 1919− 1946− Post- 1946− Post-
All Presidential Parliamentary presidential LA Non-LA 1870 1918 1945 1979 1979 1979 1979
Constitutions with executive decree powers 65.4 70.2 52.1 71.4 78.5 55.6 63.4 70.5 100.0 81.5 94.7 61.5 51.3
Decree Implementation
Effective immediately once issued 17.3 17.8 41.3 41.7 5.7 48.0 0.0 3.2 0.0 18.2 11.1 41.7 60.0
Effective following a specified period 1.1 1.7 0.0 3.3 0.8 4.0 0.0 0.0 0.0 0.0 5.6 4.2 5.0
during which an approving body can repeal it
Effective only after approval from the approving body 19.2 37.4 23.8 18.3 46.8 14.0 61.5 41.9 51.9 40.9 33.3 16.7 5.0
Not specified 27.5 37.9 38.1 43.3 42.7 26.0 34.6 51.6 44.4 36.4 44.4 25.0 30.0
Decree validity
Permanent, unless repealed 7.3 8.0 12.7 10.0 4.0 18.0 3.9 3.2 0.0 4.6 11.1 16.7 20.0
Naturally expires, unless extended 10.5 10.3 23.8 18.3 3.2 28.0 0.0 0.0 0.0 18.2 0.0 25.0 35.0
Not specified 47.0 75.3 46.0 56.7 87.9 44.0 96.2 90.3 92.6 77.3 77.8 45.8 40.0
Decree approving body

Executive 23.6 44.8 28.6 33.3 55.7 18.0 76.9 61.3 51.9 50.0 27.8 20.8 10.0
Texas Law Review
Legislature 24.8 27.0 61.9 51.7 19.4 46.0 7.7 9.7 22.2 36.4 27.8 37.5 60.0
Not specified 20.6 26.4 30.2 33.3 26.6 26.0 19.2 29.0 29.6 18.2 38.9 25.0 30.0
Executive is authorized to issue decrees . . .

That pertain to war or conflict 5.8 12.1 6.4 13.3 12.9 10.0 7.7 6.5 11.1 18.2 27.8 8.3 0.0
During states of emergency, exception, siege, or urgency 17.6 24.7 44.4 4.0 24.2 26.0 7.7 12.9 25.9 50.0 33.3 25.0 30.0
On matters of foreign policy 0.8 0.0 3.2 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
When the legislature is not in session 11.1 12.1 28.6 11.7 10.5 16.0 7.7 6.5 14.8 18.2 5.6 16.7 10.0
Left explicitly to constitutional law 6.1 9.2 7.9 8.3 10.5 6.0 0.0 9.7 14.8 13.6 16.7 12.5 0.0
Not specified 21.4 33.9 19.1 31.7 35.5 30.0 50.0 25.8 44.4 27.3 27.8 20.8 40.0
Note: Entries represent the percentage of constitutions in each category (column) that provide for a given feature (row)
[Vol. 89:XXXX
2011]
Appendix G. Constitutions that Grant Legislative Initiative to the Executive, 1789−2007

Presidential Latin American Presidential Non-LA Pres.
Semi- Pre- 1870− 1919− 1946− Post- 1946− Post-
All Presidential Parliamentary presidential LA Non-LA 1870 1918 1945 1979 1979 1979 1979
Constitution provides for:
Constitutional amendments 98.0 99.2 95.0 98.8 100.0 97.8 100.0 100.0 100.0 100.0 100.0 97.4 97.4
Budget laws 78.6 75.4 78.5 88.1 75.3 75.6 46.3 77.3 81.5 96.3 94.7 84.6 71.8
Organic laws 18.8 14.9 9.9 42.9 11.4 21.1 0.0 6.8 3.7 22.2 42.1 15.4 28.2
Finance laws 20.5 12.1 32.2 28.6 8.2 18.9 2.4 11.4 3.7 18.5 5.3 12.8 25.6
Tax laws 29.6 23.8 44.6 25.0 20.9 28.9 17.1 22.7 18.5 29.6 15.8 20.5 38.5
Spending laws 19.4 14.9 34.7 10.7 11.4 21.1 2.4 15.9 11.1 18.5 10.5 18.0 23.1
If the constitution allows for it, the executive can propose:

Constitutional amendments 43.2 34.2 36.5 79.5 22.2 55.7 4.9 2.3 25.9 29.6 89.5 65.8 52.6
Budget laws 32.9 47.1 19.0 14.9 58.8 26.5 42.1 29.4 63.6 84.6 88.9 27.3 25.0
Organic laws 8.2 10.8 0.0 8.3 11.1 10.5 0.0 0.0 0.0 0.0 25.0 0.0 18.2
Finance laws 22.6 36.7 20.5 8.3 30.8 41.2 0.0 0.0 ** 40.0 ** 40.0 50.0
Tax laws 17.9 18.6 20.4 9.5 9.1 30.8 0.0 0.0 0.0 12.5 66.7 50.0 26.7
Spending laws 35.2 48.7 26.2 22.2 61.1 36.8 ** 28.6 66.7 80.0 100.0 57.1 22.2
Latin American Presidentialism
In case of the legislature's failure to pass a proposal, the budget defaults to:
Executive's proposal 9.3 10.7 6.3 9.5 7.6 16.2 0.0 0.0 4.6 23.1 11.1 15.2 17.9
Executive's proposal, if
proposed prior to failure 12.8 11.4 11.1 27.3 12.9 5.6 0.0 0.0 7.1 27.3 12.5 0.0 14.3
Previous year's budget 33.2 32.1 28.4 41.9 30.3 35.3 5.3 32.4 31.8 42.3 33.3 45.5 28.6
Other 14.9 9.1 23.2 18.9 4.2 17.7 0.0 0.0 9.1 7.1 5.6 24.2 10.7
Note: Entries represent the percentage of constitutions in each category (column) that provide for a given feature (row)
33
Readers with comments may address them to:
Professor Tom Ginsburg

University of Chicago Law School
1111 East 60th Street
Chicago, IL 60637
tginsburg@uchicago.edu
The University of Chicago Law School
Public Law and Legal Theory Working Paper Series
For a listing of papers 1–345 please go to http://www.law.uchicago.edu/publications/papers/publiclaw.
346. Rosalind Dixon and Richard Holden, Constitutional Amendment Rules: The
Denominator Problem, May 2011
347. Rosalind Dixon, Constitutional Amendment Rules: A Comparative Perspective, May
2011
348. Rosalind Dixon, Weak-Form Judicial Review and American Exceptionalism, May 2011
349. Rosalind Dixon, Transnational Constitutionalism and Unconstitutional Constitutional
Amendments, May 2011
350. Adam B. Cox and Richard T. Holden, Reconsidering Racial and Partisan
Gerrymandering, May 2011
351. Brian Leiter, The Circumstances of Civility, May 2011
352. Brian Leiter, Naturalized Jurisprucence and American Legal Realism Revisited, May
2011
353. Lee Anne Fennell, Property and Precaution, June 2011
354. Alon Harel and Ariel Porat, Commensurability and Agency: Two Yet-to-Be-Met
Challenges for Law and Economics, June 2011
355. Bernard E. Harcourt, Radical Throught from Marx, Nietzsche, and Freud, through
Foucault, to the Present: Comments on Steven Lukes’ “In Defense of False
Consciousness,” June 2011
356. Alison L. LaCroix, Rhetoric and Reality in Early American Legal History: A Reply to
Gordon Wood, July 2011
357. Martha C. Nussbaum, Teaching Patriotism: Love and Critical Reform, July 2011
358. Shai Dothan, Judicial Tactics in the European Court of Human Rights, August 2011
359. Jonathan S. Masur and Eric A. Posner, Regulation, Unemployment, and Cost-Benefit
Analysis, August, 2011
360. Adam B. Cox and Eric A. Posner, Delegation in Immigration Law, September 2011
361. José Antonio Cheibub, Zahcary Elkins, and Tom Ginsburg, Latin American
Presidentialism in Comparative and Historical Perspective, September 2011

Cheibub-Latin America Presidentialism PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Cheibub-Latin America Presidentialism PDF

Hochgeladen von

Copyright:

Verfügbare Formate

CHICAGO

Electronic copy available at: http://ssrn.com/abstract=1899637

José Antonio Cheibub,* Zachary Elkins,**

* Professor of Political Science, University of Illinois at Urbana-Champaign.

Electronic copy available at: http://ssrn.com/abstract=1899637

parliamentary systems.7 In turn, heads of government in parliamentary

7. Thomas Weishing Huang, The President Refuses to Cohabit: Semi-Presidentialism in

Electronic copy available at: http://ssrn.com/abstract=1899637

One implication of this insight is that scholars need to explore

I. The Shadow of the U.S. Constitution

Electronic copy available at: http://ssrn.com/abstract=1899637

Venezuela‘s constitution of 1811, Mexico‘s of 1824, Argentina‘s of 1826,

Judicial review was also an American export. Many constitutions in the

constitutional replacements in Latin America40 that in theory would have

II. The U.S. Constitution as the Archetype of Presidentialism

40. See infra Appendix A.

no constitutional power of executive decree.54 Third, the President has

archetypical U.S. model of a constitutionally weak president. It may be that

III. Latin American Presidentialism

64. We thank John Ferejohn for forcing us to clarify this point.

is it part of an overall trend, if not in all constitutions, at least in presidential

systems exhibit the same level of coherence as non-presidential systems after

IV. Executive Lawmaking Power as a Signature Feature of Latin American

characterized Latin American republics since independence, as Loveman

more constrained in presidential than in parliamentary and semi-presidential

C. Constitutional Amendment and Initiation of Legislation

was proposed by the executive. Other solutions, including adopting the

V. Is Presidential Lawmaking Desirable?

seen as normatively attractive to prevent tyranny.90 We thus must ask

(2000); George Athan Billias, Introduction to AMERICAN CONSTITUTIONALISM ABROAD, supra

individuals even in pure parliamentary systems, where scholars have spoken

contemporary outcomes.99 Finally, while the legal-origins analysts

Appendix A. Available and Missing Latin American Constitutions by Decade

Nicaragua (1) (1) (1) 2 1 1 1 1 (1) 1 1 1 1 14

Appendix B. Percentage of Latin American Constitutions with

Number of constitutions: 190 63 46 29 32 20

Appendix C. Trends of Executive–Legislative Provisions in Latin

Solid line = Latin American constitutions

Appendix D. Similarity Between Constitutions with Respect to

Pre-1870 0.74 0.75 0.70

1870−1918 0.74 0.75 0.43

1919−1945 0.69 0.73 0.61 0.63 0.61

1946−1979 0.66 0.78 0.65 0.63 0.58

Post-1979 0.66 0.82 0.64 0.66 0.61

All 0.68 0.73 0.63 0.64 0.59

Body that approves emergency declaration:

Emergency can be declared:

Restrictions on emergency powers:

Decree approving body

Executive is authorized to issue decrees . . .

Appendix G. Constitutions that Grant Legislative Initiative to the Executive, 1789−2007

If the constitution allows for it, the executive can propose:

Professor Tom Ginsburg

For a listing of papers 1–345 please go to http://www.law.uchicago.edu/publications/papers/publiclaw.

Das könnte Ihnen auch gefallen